论文部分内容阅读
本文通过语言的声学特征、音律信息及语言结构的统计概率考察了蒙古语的 主要方言。分析方法包括：语音的 ＬＰＣ特征参数共振峰频率分析；方言语音识别分 析和识别音素串（口语）与文语（书面语）比较分析；利用方言对照字典，语言结构概 率统计分析；利用逐渐逼近法分段线近似基音（Ｆ０）曲线，抽出各分段线的起始频率 和斜率参数后，用判别分析法进行方言分析。分析结果表明，上述各类方法不仅直 观、符合实际而且信赖性较好。
In this paper, the main dialects of Mongolian language are investigated through the statistical characteristics of acoustic features, temperament information and language structure. Analytical methods include: LPC charac- teristics of speech, frequency analysis of formant; dialect speech recognition analysis and recognition of phonetic string (colloquial) and text (written language) comparative analysis; dialect-controlled dictionaries, statistical analysis of language structure probabilities; Line segment approximate pitch (F0) curve, each segment line out of the initial frequency and slope parameters, using discriminant analysis of dialect analysis. The analysis results show that the above methods are not only intuitive, practical and trustworthy.